public inbox for b.a.t.m.a.n@lists.open-mesh.org
 help / color / mirror / Atom feed
From: elektra <onelektra@gmx.net>
To: The list for a Better Approach To Mobile Ad-hoc Networking
	<b.a.t.m.a.n@open-mesh.net>
Subject: Re: [B.A.T.M.A.N.] a proposal
Date: Fri, 14 Sep 2007 13:51:43 +0200	[thread overview]
Message-ID: <200709141351.43182.onelektra@gmx.net> (raw)
In-Reply-To: <32455.85.41.146.90.1189767047.squirrel@krisma.oltrelinux.com>

Hi Antonio -

your suggestion sounds reasonable and is worth thinking about. However I think 
this issue will become less important with B.A.T.M.A.N.-IV since it will 
massively improve the routing metric. I think it is fair enough that a 
gateway client will unilaterally choose the gateway - the number of OGMs 
received should provide enough information for the client to choose the right 
gateway. 

cu elektra


> A client node take its decision about chosing a gateway in a unilateral
> way, that is chosing gateway on a ranking basis (or wathever else)
> checking the path "node-->gateway" or "node-->hop-->gateway" (also if the
> second is not always preferred, chosing a poor link "node-->gateway"
> rather then best links "node-->hop-->gateway").
>
> Presuming the node has a bouquet of canditate gateways: we can think to
> use an udp packet polling the candidate gateways and asking them "how you
> see me?" so the the node itself can obtain a path evaluation computed at
> gateway side and not only at node side.
> For example
> presuming node A can chose among gateways G1 and G2 (the candidate
> gateways) it will make its choice upon path evaluation:
>
> A --> hops --> G1
> A --> hops --> G1
> where [hops] can be also 0 (no hops between A and Gn)
>
> and computing OGMs coming from neighboors.
>
> Well, node A can send two udp packet (same format as OGM but different
> op-code in payload||different dest port) to G1 and G2 asking to send their
> path evaluation like "G1 --> hops --> A" evaluation (received from
> candidate G1) and "G2 --> hops --> A" evaluation (received from candidate
> G2). In this way node A has a more clear vision about the entire path A
> <--> G_candidate and can take a more serious decision.
>
> Using udp requests we introduce low latency than tcp and with a separate
> thread on gateway (with a poor resources consumpion) we can manage this
> task without effetcing the main.
>
> What are you thoughts about?
>
> --
> Antonio
>
> _______________________________________________
> B.A.T.M.A.N mailing list
> B.A.T.M.A.N@open-mesh.net
> https://list.open-mesh.net/mm/listinfo/b.a.t.m.a.n



  reply	other threads:[~2007-09-14 11:51 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-09-13  6:08 [B.A.T.M.A.N.] translation of batman-adv 唐鼎
2007-09-13 11:12 ` [B.A.T.M.A.N.] terra-net (Mesh im Handy) thomasasta
2007-09-14  6:46 ` [B.A.T.M.A.N.] translation of batman-adv Simon Wunderlich
2007-09-14  7:11   ` 唐鼎
2007-09-14  8:03     ` Simon Wunderlich
2007-09-14 10:30       ` ??
2007-09-14 10:50         ` [B.A.T.M.A.N.] a proposal a.anselmi
2007-09-14 11:51           ` elektra [this message]
2007-09-14 11:32         ` [B.A.T.M.A.N.] translation of batman-adv Simon Wunderlich

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=200709141351.43182.onelektra@gmx.net \
    --to=onelektra@gmx.net \
    --cc=b.a.t.m.a.n@open-mesh.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox